Maximilian Algehed

The reason for making this change is the change in #383 that means that `maxSuccess` is never really _max_ success (it wasn't before either - but now it's even more...

> @MaximilianAlgehed https://hackage-search.serokell.io/ is better corpus to search through Thank you! I'll do a more thorough search there.

There are about 50 odd packages on hackage that use `withMaxSuccess` that we have to care about (i.e. aren't mentioning it only in a comment).

@nick8325 unless there is some reason for this that I can't make sense of I'm going to add instances of `CoArbitrary` and `Function` for the types in `Test.QuickCheck.Modifiers`.

This is the intended behaviour of `Gen` as the lazyness allows you to be a bit more fast and loose. One option to make working with `discard` slightly more ergonomic...

Well, `quickcheck-instances` provides orphan instances for things in base that should have been provided by `QuickCheck` from the start. Updating the version bounds on `quickcheck-instances` is trivial and this avoids...

@nick8325 thoughts?

My take on this is that `base` changing makes things awkward and there is no good way around that. Most people will use `QuickCheck` on a recent GHC version and...

Fine, longer build times then.

Considering we released a new version of QuickCheck a while ago and `quickcheck-instances` hasn't kept up (I even made a PR that was summarily rejected to bump the versions), I'm...